For bug reports or feature requests concerning my tools on Labs, please use BitBucket.
It helps me keeping track of all the requests. Beats spreading them over half a dozen talk pages ;-)


is there any chance to block anonymous user from reverting images? it seems a script can compute the link, which reverts an image to an old version. something like this is going on on de.wikipedia.

-- 16:45, 23 Jun 2004 (UTC)

HTML to wiki converter edit

I'd like to thank you for this great tool, it makes my life a zillion times easier. Thanks again and keep up the great work! Halibutt

- Kannst Du den Converter vielleicht auch auf WikiMedia stellen - die Uni Bonn wird es vielleicht nicht ewig auf ihren Rechnern laufen lassen... - Danke Sebastian

Enhanced Recent Changes as Default edit

Magnus amazing work!. I am currently running MediaWiki version 1.5Beta3 and was wondering how can I set as default the Enhanced recent changes for all users when they register their account. While still leave the option available for users preferences to uncheck it. I went briefly over the DefaultSettings.php but couldn't find the parameter related to this setting.

Thanks! --Martin 21:12, 11 August 2005 (UTC)Reply

Tables / col edit

what do you think? --grin 10:22, 3 Dec 2004 (UTC)

RfA revisited edit

Hi Magnus,

Your adminship is up for review (along with everyone else who became an admin in the same n-month period). Do you still want adminship on meta?

Rock on, +sj+ 01:10, 19 Jan 2005 (UTC)

Download the client edit

Hi Magnus, am I stupid? Where can I get the software? I've tried your download-link but it was not reachable!? Dieter 15:41, 25 Jan 2005 (UTC)

Article validation feature edit

See Talk:MediaWiki feature list#Article validation feature. See Talk:Article validation feature#Where is the code?. Brianjd | Why restrict HTML? | 06:07, 2005 Mar 13 (UTC)

Magnus, please check En validation topics - User:Beland has taken most of the current suggestions and put them together into a form that I think will work very nicely (En_validation_topics#Consolidated_proposal). There's a couple of questions I've asked you on that page as well - David Gerard 22:59, 30 May 2005 (UTC)Reply

Magnus, I have a few requests for this feature, which taken together, I believe, will substantially improve it:

Article (stable-version) - URL:""
  • firstly, the review is worthless to a reader unless the reader is somehow affected by it. A reader should be able to see the page with the highest rating.
    • I believe there should be another tab on the top for this, perhaps "stable", or replace article with "draft" and make the new tab "article". "article" is version with highest rating, "draft" is latest version.
    • I believe the stable version should be the default for non-logged-in users, and the draft for logged-in users.
      • (low priority) registered users should be able to choose the default in their preferences.
Click where a thumb would be to flip to thumbs up(approve), click again to flip to thumbs down(disapprove). overall rating is shown in horizontal bar. Current stable version (highest rated) is highlighted. (sorry about the bad image editing)
  • there should be a horizontal bar in the history, for each version, showing it's rating.
    • (low priority) it would be nice to be able to vote via this bar
    • (low priority) it would be nice to have a similiar feature in diffs, and be able to vote from there
  • secondly, this means that a really old version with a high rating might be the "stable" version, while a newer, lower rated, but better article, is not, there are solutions to this
    • logarithmic decay of effective ratings
    • maximum number of revisions a user can rate on an article at a time - rating an revision that you haven't rated removes your rating of the rating you made the longest time ago (not neccessarily the oldest revision)
  • ability to filter out revisions in the history that are below a specifiable rating threshold.

That's the jist of it. Feel free to discuss. Thanks. Kevin Baastalk 01:15, 4 April 2006 (UTC)Reply

Historical atlas edit

Hi Magnus,

I want to help you creating an online historical atlas. At the moment atlases in book form are better then the online versions. But internet has such huge possibilities, that the online maps should become much better. Maps could be clickable giving links to other maps and data in text. Maps could give detailed information on areas in small intervals in time. Alternative maps on an area could be made, when there is a lack of sources.

I'm studying history in Leiden, Holland and no little about software. But i have been looking on historical maps since my childhood. In the 1990s an online historical atlas was made, called Centennia, but the information on it is limited (Europe and Middle East from 1000 untill now) and it is not free, it costs $89. Maps in modern historical atlases are mainly derived from German historical atlases from the 19th century. German atlases are the most accurate, but American atlases give a good overview.--Daan 14:02, 27 August 2005 (UTC)Reply

Google Summer of Code edit

Hi, I'm sorry to disturb you, I know you have been working on a wiki2xml export tool. I would like to know who I have to contact about the project of Export and how it works :) Thank you for your response.


missing image tool edit

suggestion: add *disambig* to the filter instead of adding many images with that word to the list. drini 01:30, 7 October 2006 (UTC)Reply

Likewise *Nuvola* and *Noia* which are some widely used icon sets for disambigs and stubs. drini 01:47, 7 October 2006 (UTC)Reply

Perhaps Crystal*.png and Qsicon*.png ? drini 01:53, 7 October 2006 (UTC)Reply

missing image tool redux edit

is there a way for the script work on a given list ionstead of a random choice? For instance, in an article about a Jules Verne nove, script foudn a Jules Verne picture, but then I'd like to search for a frigate or other related items, but I don't know how to feedback the script with the list of articles I want pictures about. drini 17:06, 9 October 2006 (UTC)Reply

Annoying you again edit

I was curious about what do colors represent on the image missing list. drini 21:38, 9 October 2006 (UTC)Reply

One problem edit

Hi :-) I find really useful your missing images tools :-) But i have a small problem, in articles with names with strange chars it doesn't seems to work; For example the article ca:Nombre d'Avogadro has an image and interwiki links, but the search doen't find it... Result number 7 with search parametres;

  • Check in: ca.wikipedia
  • Category : Química
  • Subcat depth : 0

Thanks :-) --Joanjoc 15:20, 5 November 2006 (UTC)Reply

Thanks for your rapid response :-) Another minor problem, runing the same query as above but with a subcat depth of 1, appears some redirects; ca:Elements químics and ca:Química física. The first page only contains the redirect code without images nor categories, but the destination page has iw not reported by your magnific tool... --Joanjoc 20:43, 8 November 2006 (UTC)Reply

Macedonian for CatScan2 edit

Hello Magnus, I added the complete Macedonian translation (mk) of CatScan2 on CatScan2/Interface and the manual (CatScan2/mk) and I was wondering if you could tell me when we can expect to see it implemented. Thanks a lot! --B. Jankuloski 02:31, 7 July 2010 (UTC)Reply

It was "implemented" the moment you put the translation on the interface page. If something's not working, please tell me. --Magnus Manske 07:58, 7 July 2010 (UTC)Reply

Portuguese Makeref/Toolserver edit

First, my appologies for making my request by this talk page instead of by system JIRA, but I don't know how these system works.

I created a portuguese version of makeref (here and here), but for some reason the option "Um artigo num jornal académico ou num periódico" is not working.

Can you explain what I am doing wrong?

Thanks in advice--MiguelMadeira 16:34, 27 December 2010 (UTC)Reply

No more help needed - I discovered what the problem is--MiguelMadeira 18:06, 27 December 2010 (UTC)Reply

Main categories edit

Hi Magnus - Thank you very much for the Catscan tool, it is so very useful! I am currently doing some academic research on English Wikipedia article pages and I would like to categorize them at the Main category level. For this, I was trying to use with categories such as "education", "business" and depth 50 (or so) to generate complete lists of all the articles in each category (so, 10 lists). Obviously some articles will be in multiple lists.

However I suspect this query is difficult for the toolserver because of the sheer amount of nodes it needs to explore in the category tree. Is there any way you can help me get those 10 lists?

Thank you so much!!

Best, Andreea

email agorbatai(at) hbs (dot) edu)

commonshelper2 edit

Hello. There was a small question on local village pump about commonshelper's work recently and I memorised that is still doesn't work for Bug is still opened but you've proposed to update commonshelperII instead of trying to fix commonshelper. I've updated necessary data in April but I suppose CHII wasn't updated according to this info. Is it possible to fix this issue? can't transfer images to Commons since the beginning of 2011 if I am not wrong... :( rubin16 14:02, 14 August 2011 (UTC)Reply

GLAMorous and Atlassian problems edit

The whole Atlassian thing is a complete fiasco -- I spent at least 15 minutes at that site, and never managed to login (at one point, it told me I was logged in, but I actually wasn't).

In any case, GLAMorous has problems, in that:
1) The "Copy the URL of this link to return to this view. This data is also available in XML format." links do not work if you're searching by user. I had to manually munge the URl to
2) An alternative "slightly detailed" mode is really needed. When I clicked on the second of the URLs above, the download stopped after 3.81 megabytes was downloaded, and the data was still in the middle of reporting on the first image (out of 1339) on the list, File:Flag_of_Turkey.svg. I think people would be interested in a view which did NOT report on individual page uses, so that they could see which of their images were most used in a reasonably-sized overview.
Thanks! -- AnonMoos 16:19, 22 December 2011 (UTC)Reply

XSS vulnerability edit

I found a cross-site scripting (XSS) vulnerability in CatScan. Example:;%3C/script%3E (will open an alert box). I don't know how to use JIRA, but it looks like I'll need a toolserver account (I don't have one) and/or to give out personal info, so I'm writing this here. πr2 (tc) 16:28, 28 March 2012 (UTC)Reply

Several (all?) of your other tools are vulnerable as well. (could it be in some shared code?) πr2 (tc) 16:58, 28 March 2012 (UTC)Reply
@PiRSquared17 I had just sent Magnus an email regarding a similar vulnerability in PetScan. Kins of obnoxious that issues like this can't be fixed in such a long time (10 years?). ~~~~
User:1234qwer1234qwer4 (talk)
21:05, 11 January 2022 (UTC)Reply
(For reference, the above has been fixed after being submitted as Phabricator tasks.) ~~~~
User:1234qwer1234qwer4 (talk)
19:26, 5 June 2022 (UTC)Reply

TreeViews yearly? edit

Hi Magnus!

Could TreeViews be expanded to include yearly statistics, or are those better found elsewhere? --Palnatoke (talk) 13:45, 9 May 2012 (UTC)Reply

Catscan - thanks! edit

Thanks for the Catscan fix and advice! Allens (talk) 16:26, 22 May 2012 (UTC)Reply

Cambridge fresher's fair edit

Hello Magnus,

I've bid a stall on behalf of the Cambridge University Wikipedia Society at the Cambridge University fresher's fair, 2-3 October. If you are around, it would be great if you can come and help by staffing the stall for a few hours - having one of the inventors of MediaWiki man the stall will certainly be a boost!

The fair will run, roughly, from 9am to 7pm on the 2nd, and 9am to 4pm on the 3rd; details are yet to be confirmed.[1] You certainly won't be expected to stay the whole day unless you really want to! It'll be a sign-up stall for the campus Wikipedia Society, and we'll give out Wikimedia freebies at the event to promote awareness for Wikipedia-editing and Wikimedia UK.

Please contact me if you're interested in helping, even if it's just a few hours. Thanks! (Apologies for double-posting if you've seen this message on wikimediauk-l already!) Deryck C. 22:30, 5 September 2012 (UTC)Reply

Wikipedia DVD/offline edit

Hi Magnus, I saw your contribution on OK you may be the guy to give me some advice. I am teaching in a country riddled not only by censorship but also by slow internet. We discussed at our expat school if there is an option to install wikipedia directly onto our school server so that we can have the kids work with it via school net but without being subject to the frequent interruptions. Besides that I have a very strong interest in some advice how you practically produced that German DVD you mentioned. If you think you can help me kindly leave me a line at --Kipala (talk) 05:51, 25 October 2012 (UTC)Reply

Cambridge meetup March 2013 edit

Is that you? IP resolves to Sanger. Deryck C. 21:25, 20 February 2013 (UTC)Reply

Yes, thanks! Wikimedia sites keep logging me out randomly :-( --Magnus Manske (talk) 22:37, 20 February 2013 (UTC)Reply

Reasonator edit

Do you plan to create the page Reasonator? πr2 (t • c) 19:38, 8 April 2013 (UTC)Reply

Maybe, eventually, but feel free to go ahead! --Magnus Manske (talk) 21:25, 8 April 2013 (UTC)Reply
Created. You can check it or add more details... I don't have time to check it. Would you like to upload File:Reasonator screenshot Bach.png (like you've done for File:GeneaWiki screenshot Bach.png)? πr2 (t • c) 05:00, 9 April 2013 (UTC)Reply

Requests for BaGLAMa edit

Hi Magnus!

I was wondering if you could be so kind to help me by adding the categories that I have listed here to BaGLAMa? It would really help WMSE as some of the GLAMs involved want some statistics to brag about :-).


John Andersson (WMSE) (talk) 14:47, 22 June 2013 (UTC)Reply

Added. They will show up on the next run after the current one. Couplea days. --Magnus Manske (talk) 15:07, 22 June 2013 (UTC)Reply
That's great! Thank you! Could you also add It would be interesting to see the effect of this project. Best, John Andersson (WMSE) (talk) 20:37, 22 June 2013 (UTC)Reply

Talk:Flickr2commons#Mass-edit_title edit

FYI (thanks so much for this wonderful tool). --Nemo 08:14, 18 August 2013 (UTC) P.s.: The notice at top is outdated, if there's a better place where to track labs tools reports please tell me and I'll add notes to the tools' pages here on Meta too.Reply

Reasonator not working edit

Dear Magnus, the Reasonator stopped working this week. Could you please have a look? Perhaps this has something to do with the Wikidata change Daniel Kinzler announced this week. Longbow4u (talk) 09:56, 14 September 2013 (UTC)Reply

Yup, API upgrade broke it. Should be fixed now. Thanks for alerting me. --Magnus Manske (talk) 11:31, 14 September 2013 (UTC)Reply
Thank you for the prompt fix! It is a great tool. Longbow4u (talk) 12:05, 15 September 2013 (UTC)Reply

CatScan2 bug edit

(No component named CatScan2 in JIRA) Please use database suffix "wiki" for project "wikimedia" in CatScan2 in tool labs. The original error is Could not connect to incubatorwikimedia.labsdb : Unknown MySQL server host 'incubatorwikimedia.labsdb'(115) --Zhuyifei1999 (talk) 08:16, 7 October 2013 (UTC)Reply

Should now work for incubator. --Magnus Manske (talk) 08:32, 7 October 2013 (UTC)Reply
Thanks --Zhuyifei1999 (talk) 23:24, 7 October 2013 (UTC)Reply
I just got Could not connect to roa-tarawiki.labsdb : Unknown MySQL server host 'roa-tarawiki.labsdb' (115) instead. Thanks. --Elitre (WMF) (talk) 16:04, 26 November 2013 (UTC)Reply
Fixed.--Magnus Manske (talk) 16:25, 26 November 2013 (UTC)Reply

CommonsHelper and the Afrikaans wiki edit

Hi Magnus, I tried to configure the Data af.wikipedia page so that CommonsHelper2 could transfer files (such as this one) from the Afrikaans Wikipedia, but after countless attempts I still can't get it to work properly. Could you please have a look? Thanks, Underlying lk (talk) 17:00, 2 December 2013 (UTC)Reply

Flickr2Commons funktioniert nicht edit

Hallo, schaust du bitte einmal hierher: Talk:Flickr2commons#Transfer_failed_:_null. Danke! darkweasel94 (talk) 10:40, 10 March 2014 (UTC)Reply

MakeRef edit

I am hoping you can spare a few moments to rename "MakeRef/mr" to "MakeRef/Mr" and undelete "MakeRef/Fr" (if it contains useful information, that is). Thanks! -- 18:11, 11 April 2014 (UTC)Reply

Please also consider these changes. -- 18:11, 11 April 2014 (UTC)Reply

Did the linked changed. Moved the page. Can't resurrect "MakeRef/Fr", as I'm not an admin here. --Magnus Manske (talk) 10:06, 16 April 2014 (UTC)Reply

Five out of six isn't bad. Many thanks for attending to this. -- 17:07, 16 April 2014 (UTC)Reply

Current CatScan bug edit

When trying to get a category scan (e.g this one in wiki format, the page names are missing. Can you please fix that? עוד מישהו Od Mishehu 10:27, 9 November 2014 (UTC)Reply

Fixed. Note that I am rarely on meta, so bug reports here can go unnoticed for a long time. --Magnus Manske (talk) 09:00, 13 November 2014 (UTC)Reply

Category tree edit

Hello. Could you tell me, please, is there any possibility for scanning a Wikipedia on some language (37788 categories) and return for each category an inverse tree of parent categories up to some level? For example, for level 2, category a is in b and in c, category b is in d and e, category c is in e, or something like that? Thank you very much in advance, IKhitron (talk) 18:53, 2 December 2014 (UTC)Reply

There's no ready-made tool, but it's certainly possible. Which language were you thinking of, and what output format? --Magnus Manske (talk) 19:04, 2 December 2014 (UTC)Reply
Thank for your quick answer. It's the he Wikipedia (so, unfortunately, non ASCII names can be an issue). Output format is not so important, as long as it will be compact (meaning the tree for each category is continuous). If you just add some print function and it is not a problem to add any format, I prefer CSV. But I do not want to bother you. Thank you very much, IKhitron (talk) 19:26, 2 December 2014 (UTC)Reply

Broken Link to Missing Images Tool edit

Same here:

CatScan usage for bots edit

Hello Magnus,

I'm a little bit troubled with the usage of your cool tool with a bot. How I see it my bot isn't able to tell your tool, which output format is to use. Maybe you can give me a hint to a documentation, which goes further than the gui.

Regards --THE IT (talk) 12:12, 29 January 2015 (UTC)Reply

You can use "format=X", with X one of html, csv, tsv, wiki, json. For example, JSON. --Magnus Manske (talk) 13:51, 29 January 2015 (UTC)Reply
Thanks a lot. That’s a step further ;-). A brief overview over all options (maybe later when the tool is finished) would be very helpful. If I hit further problems ... now I have found your xmpp-Adress. Again many thanks. --THE IT (talk) 17:31, 4 February 2015 (UTC)Reply
Hello Magnus, sadly there is further struggle. When I try to acces the link with a python-script (via the Post method) the access is forbidden (403) with the reason 'Requests must have a user agent'. Can you give me a hint about my mistake? Regards --THE IT (talk) 17:59, 8 February 2015 (UTC)Reply
I don't make the web server (lighttpd), but it seems you need to specify a user agent? Any should do, just needs to be present. --Magnus Manske (talk) 10:48, 4 March 2015 (UTC)Reply

Not in the other language edit

Hallo Magnus, Entschuldigung für die Störung; ich hätte eine kleine Frage: Kann dein Not-in-the-other-language Tool auch für Artikeln zwischen zwei sprachlichen Versionen von Wikivoyage benutzt werden oder nur zwischen Wikipedia Artikeln? Danke im Voraus. Grüße --Nastoshka (talk) 09:40, 6 March 2015 (UTC)Reply

3D SVG edit

You have new messages
Hello, Magnus Manske. You have new messages at [[User talk:Meetup/Cambridge/26|User talk:Meetup/Cambridge/26]].
You can remove this notice at any time by removing the {{Talkback}} or {{Tb}} template.

Please fill out our Inspire campaign survey edit


Thank you for participating in the Wikimedia Inspire campaign during March 2015!

Please take our short survey and share your experience during the campaign.

Many thanks,

Jmorgan (WMF) (talk), on behalf of the IdeaLab team.

23:34, 13 April 2015 (UTC)

This message was delivered automatically to Inspire campaign participants. To unsubscribe from any future IdeaLab reminders, remove your name from this list

Geograph2commons possible bug edit

Dear Magnus, I believe the very wonderful Geograph2Commons tool is one of yours? If so I have a possible bug. Most images can be uploaded or tell you they have already been uploaded. But some images fail with a message of "Warning" and nothing more specific. This being an example, as far as I can see it hasn't been uploaded already and has a compatible license, but if I've missed something would it be possible for the warning to say why images like this don't upload? Many thanks WereSpielChequers (talk) 09:01, 21 May 2015 (UTC)Reply

Links sind tot edit

Die Verlinkungen oben im roten Kasten auf dieser Seite führen ins Leere. Aber den Bug den ich eigentlich melden wollte: Wenn man bei "Wiki" als Format auswählt, werden Dateien in voller Auflösung anstatt als Link dargestellt. (Es fehlt der Doppelpunkt)

    ist : [[File:...
    soll: [[:File:...

Danke, --McZusatz (talk) 12:04, 24 May 2015 (UTC)Reply

Ich würde auch gerne einen Bug melden. Wo genau muss man das tun? --THE IT (talk) 09:24, 13 August 2015 (UTC)Reply
Hier, auf meiner Deutschen oder Englischen Benutzerseite, oder (bevorzugt) hier, wenn das tool da in der Liste ist. --Magnus Manske (talk) 10:02, 13 August 2015 (UTC)Reply

Foto Haus Prinzipalmarkt 28, Münster edit

Auf der Wikipediaseite Joseph Heinrich Coppenrath wurde das Foto Haus am Prinzipalmarkt 28, Münster von CommonsDelinker gelöscht:

Münster, Prinzipalmarkt 28: Buchhandlung Coppenrath, 1933

Begründung: Zusammenfassung des Bearbeiters: Med1743.jpg entfernt, auf Commons von

JuTa gelöscht. Grund: No license since 24 June 2015

Dieses Foto ist 1933 aufgenommen das sind 82 Jahre her. Es gibt überhaupt kein Copyright darauf. Das sagte mir auch der Coppenrath Verlag. Was also habe ich nicht richtig gemacht? oder, besser wäre die Löschung rückgängig zu machen. --12weev (talk) 17:13, 6 July 2015 (UTC)Reply

Der CommonsDelinker entfernt nur den Eintrag im Wikitext; das Bild wurde schon vorher gelöscht. Warum, da musst Du schon JuTa fragen. --Magnus Manske (talk) 18:33, 6 July 2015 (UTC)Reply

BaGLAMa 2 request edit

Hi Magnus, Would it be possible to add Category:Media_from_the_National_Liberation_Museum_1944-1945 to BaGLAMa 2? Best, --AWossink (talk) 09:56, 15 September 2015 (UTC)Reply

Done.--Magnus Manske (talk) 14:29, 15 September 2015 (UTC)Reply
Awesome. Thanks! --AWossink (talk) 08:14, 17 September 2015 (UTC)Reply

Unaproved bot edit

Please, pay attention to recommendations on this page and request bot flag for ListeriaBot on Portuguese Wikipedia. Unaproved bots are not allowed per local policy.

Let me know if you need any help. Kind regards.—Teles «Talk to me ˱C L @ S˲» 07:50, 2 October 2015 (UTC)Reply

Could you kindly open the request for Portuguese Wikipedia? You can use English language as well. Thanks.—Teles «Talk to me ˱C L @ S˲» 22:10, 22 January 2016 (UTC)Reply
Your bot has been blocked on Please, place a local request in case you want to use it.—Teles «Talk to me ˱C L @ S˲» 03:39, 5 February 2016 (UTC)Reply

Thanks for the endorsement! edit

Hi @Magnus,

I would like to thank you for your endorsement on the StrepHit IEG proposal, it is really precious!


--Hjfocs (talk) 15:35, 7 October 2015 (UTC)Reply

geograph2commons bug edit

I'm experiencing a bug trying to use geograph2commons. Certain images return "ERROR: Warning" with no further information. For example:

Other images seem to work ok, but some don't and I can't work out why these particular ones fail. -mattbuck (Talk) 15:06, 9 November 2015 (UTC)Reply

Copied to [3]. -mattbuck (Talk) 15:07, 9 November 2015 (UTC)Reply
Could it be that the apostrophe in the name of the photographer is a character that is not found in the basic ASCII character set? --Schlosser67 (talk) 12:58, 3 January 2017 (UTC)Reply

Flickr2Commons edit

Hi Magnus! Firstly, congrats for your huge history of support inside Wikimedia projects! :) So, I would like to upload these great pictures from Flickr to Commons, but Flickr2Commons doesn't allow me because they're mostly bad named (just number sequences). Is there a way to automatic set a prefix to the pictures, just like "Alexander Baranov - " or something like that? Other suggestion? Regards, Sturm (talk) 19:09, 17 November 2015 (UTC)Reply

Yes there is! Once the images are loaded in flickr2commons, there is a green-ish button "Prefix selected named", which will do what you want. It's next to the "Select all" and "Deselect all" buttons. --Magnus Manske (talk) 19:30, 17 November 2015 (UTC)Reply
Found it! Great! :) But now I found that the prefix is inserted always with a double space, even if I do not insert any, like: "File by Alexander Baranov - (19411107842).jpg". Is there any problem to upload like this? Is there a way to avoid the double space? Hint: its between "...anov - (19..."; the double space is just visible in the wikimarkup! ;) Sturm (talk) 21:10, 17 November 2015 (UTC)Reply
Solve it using this basic tool, ;) Sturm (talk) 23:05, 17 November 2015 (UTC)Reply
Hummm, this Flickr account uses CC0 1.0 Universal (CC0 1.0), but Flickr2Commons tells me that "No suitable photos found. Maybe they are not under a free license?". I think that it's the case of add this license to do the list of acceptable licenses of the tool, nope? Sturm (talk) 08:56, 21 November 2015 (UTC)Reply
Try it now. --Magnus Manske (talk) 15:40, 21 November 2015 (UTC)Reply
Still not working, I've got the same message. Just to let you know if I am really doing it properly, I'm trying do it assuming that its ID is 135941882@N05. Sturm (talk) 16:43, 21 November 2015 (UTC)Reply
Besides, I drop a few "tip lines" on this last days... Sturm (talk) 16:50, 21 November 2015 (UTC)Reply
I just transferd this with flickr2sommons, from your example user. No problems. Maybe force-reload flickr2commons once to clear the cache? --Magnus Manske (talk) 18:44, 21 November 2015 (UTC)Reply

Just amazed with your tool! Congrats again! :) But another question: When I load an album with over 600 pictures, Flickr2Commons don't show thumbnails in order to avoid problems related to memory use of the browser. But, many times its very important to see the thumbnails in order to select what pictures are reasonable to upload to the project. So, I would like to suggest a feature that turns able to show the thumbnails of a large album in sets of a maximum of 600 files (like "Page 1", "Page 2"...). What do you think? And finally, any chance of a Picasa equivalent?!?!? :D Regards, Sturm (talk) 18:15, 6 December 2015 (UTC)Reply

Flickr2Commons is able to identify the files of this album as free licensed only when run and uploaded one by one, but not when I try to run the album set. Sturm (talk) 08:09, 9 December 2015 (UTC)Reply

OAuth approval edit

Hi Magnus,

I would like to request the approval of my OAuth application Kaspar [1.0]. I already did edits with it on Wikidata. Do I have to provide more information? How does this process work? -- T.seppelt (talk) 23:36, 21 November 2015 (UTC)Reply

What future IdeaLab campaigns would you like to see? edit


Hi there,

I’m Jethro, and I’m seeking your help in deciding topics for new IdeaLab campaigns that could be run starting next year. These campaigns aim to bring in proposals and solutions from communities that address a need or problem in Wikimedia projects. I'm interested in hearing your preferences and ideas for campaign topics!

Here’s how to participate:

Take care,

I JethroBT (WMF), Community Resources, Wikimedia Foundation. 03:34, 5 December 2015 (UTC)Reply

Ilse Kokula edit

I am very pleased and gratefull for your help - thank you, very much. Would it be possible to add that the pictue is from 2014? Ilse Kokula, 2014? --Sarita98 (talk) 22:25, 29 December 2015 (UTC)Reply

ListeriaBot at frwiki edit


Please request the bot flag for ListeriaBot on frwiki. It can be done at fr:Wikipédia:Bot/Statut.


--Hercule (talk) 13:11, 21 January 2016 (UTC)Reply

@Hercule: Could you do that for Magnus ? I guess that if he had to do that for every wiki, 200+ time, it would be quite tedious. I would have done that myself but I can't right now. TomT0m (talk) 13:23, 21 January 2016 (UTC)Reply
@Hercule: I tried, but the template is broken, or something. I don't speak French. --Magnus Manske (talk) 15:41, 21 January 2016 (UTC)Reply
I changed few thinks. Please change in the request :
  • PURPOSE_OF_THE_BOT : explain the purpose of your bot
  • USED_SCRIPT : the script you use, if any (like Pywikipedia)
  • ALREADY_USED_ON_(WITH_BOT_STATUS) : the major wikipedias where you have the bot flag. You can also indicate if it has the global bot flag
  • ALREADY_USED_ON_(WITHOUT_BOT_STATUS) : the major wikipedias where you edit without the bot flag (if any)
On the section "Remarques" you can indicate if you have another bot running on frwiki
--Hercule (talk) 15:52, 21 January 2016 (UTC)Reply
Lively discussion: w:fr:Wikipédia:Bot/Statut#ListeriaBot. Looks like the opposers are long-time Wikidata haters: w:fr:Wikipédia:Sondage/Wikidata. Nemo 11:14, 22 January 2016 (UTC)Reply
@Nemo bis:Lol, we are haters ? Thank you for your word. As you can read, the talks about Wikidata on fr.wikipedia have started in july 2015, since almost no problem has been solved... Looks like the supports are long-time Wikidata incompetents. --Consulnico (talk) 11:32, 22 January 2016 (UTC)Reply
I can't read French, but we have the haters on German Wikipedia as well, not-made-here syndrome and all. --Magnus Manske (talk) 11:54, 22 January 2016 (UTC)Reply

Future IdeaLab Campaigns results edit


Last December, I invited you to help determine future ideaLab campaigns by submitting and voting on different possible topics. I'm happy to announce the results of your participation, and encourage you to review them and our next steps for implementing those campaigns this year. Thank you to everyone who volunteered time to participate and submit ideas.

With great thanks,

I JethroBT (WMF), Community Resources, Wikimedia Foundation. 23:56, 26 January 2016 (UTC)Reply

ListeriaBot @ etwiki edit

ListeriaBot regular updating of pages seems not to work in etwiki, for example in list --WikedKentaur (talk) 18:04, 27 January 2016 (UTC)Reply

Should be fixed now. --Magnus Manske (talk) 19:27, 27 January 2016 (UTC)Reply

Thanks. And one more request. We use coordinate template compatible with German wiki Template:Coordinate. So calls like


will work in etwiki and enwiki template Coord calls won't work in etwiki. Could you switch ListeriaBot coordinate template from Coord to Coordinate for etwiki. --WikedKentaur (talk) 18:09, 30 January 2016 (UTC)Reply

frwiki edit


after some problematic uses, ListeriaBot has been denied bot status on the French wikipedia. I don't know how this is supposed to work now. Is it possible for you to block it from editing on the French wikipedia ? Thanks, Jean-Jacques Georges (talk) 15:58, 4 February 2016 (UTC)Reply

I'm not agree with Jean-Jacques Georges, the majority have the yes for your bot (51%, 20% against in the main, 29% against), but not enough for the "flag bot", the best is to continue using your bot but not on the main . Sebk (talk) 16:15, 4 February 2016 (UTC)Reply
Actually there is a consensus to disable it in article namespace on the template talk page [4], but there is no discussion showing a consensus to fully disable it. If you can throttle it to avoid flooding RC, it should be fine. Akeron (talk) 16:19, 4 February 2016 (UTC)Reply
Well, your bot needed 75%, which it did not obtain in any case, so normally it shouldn't be able to edit anywhere on frwiki. Jean-Jacques Georges (talk) 16:32, 4 February 2016 (UTC)Reply
Jean-Jacques Georges. you are wrong... we voted for the "flag bot" not for the bot and you know it. Sebk (talk) 16:40, 4 February 2016 (UTC)Reply

Geograph upload - no text edit

Hi The Geograph upload tool isn't uploading the text

eg recent upload of


Can it capture the text Our Lady of Lourdes & St Peter Chanel Catholic Church, No.119 Cottingham Road, known locally as the Marist Church. The Society of Mary (Marist Fathers and Brothers) is an international congregation of priests and brothers which has its headquarters in Rome. A temporary church was erected in Cottingham Road by the Marist Fathers in 1925 and replaced by the present church in 1957 (Architect: John Houghton).

? Thanks Xiiophen (talk) 15:39, 6 February 2016 (UTC)Reply

Improve wikishootme with specific wikimetric targeting edit


May I disturb you? Mayne you remember me about the wiki needs picture grant. I am not here to disturb you with those details, I am only doing 20% of the work there, so if there is any curious question for you that would be for AlessioMela to make. I am here because as stated in the grant our aim is to create cooperative tools, not comptitive tools. Since I am in charge of the "social" aspect of the operation I show image tools to newbies at wikimedian meetings, ask for feedback etc

Well, long story short: wikishootme is great and is a good starting point. We are missing users with spcific tematic interest on wikidata and fixing information about the territories they're living is a nice introduction, it kinda works. Wikishootme suggest very simple tasks, less complicated than finding images on flickr and check copyroght tags etc. So it is nice, plus is a very concrete way to introduce to geo-coordinate and I have to confess that this step always takes a good 15 minutes in wikimedia classes for newbies. it is good, no doubt about it. That's why I show it with pleasure to newbies for feedback since our tool is not 100% operational.

if you remember I politely suggested that one problem of wikishootme is that it has a still considerable amount of false negatives and positives. False negatives are created by the lack of coordinates information, false positives by the slow update of the image properties on wikidata. Newbies all confirmed this impression referring to their area whilst exploring it. this means it will take some years for the tool to be efficently used and I would like to to fix the gap.

So, this year I've started to spam users about wikishootme and how to improve the data it relies on. A 20% of them seems responsive. it is not great but it is something and we can perform this strategy globally if we want.

The point is: we can do more. I can use some basic wikimetrics I know in order to select automatically "trusted local users" (as I did with itWikipedia, I just rememeber their names from many other initiatives) than filter them by a minimal acitvity on wikidata and spam a message pointing out the importance to check coordinates and image properties. It's not going to fill 100% the gap but it shouldn't take so much time compared to the results. I think I can find some users who are willing to help me on this technical part, if you just want to provide a simple supervision.

Would you support the idea? I was thinking of proposing a grant but we can start to target a group of users of one local wikipedia I can speak the language. Like the German one, as a test. My data so far with itWiki users are not complete, I need a full scale test before proposing a grant for all the platforms, so I know how to refine it. i don't want to waste WMF money... Plus italian wikipedia is mainly based on Italian territory where the copyright laws are a mess. German-speaking cuntries show a very localized activities but in area where there is for example freedom of panorama, so a spam message about finding local images shouldn't be handled with too much care,

Would you like the idea?--Alexmar983 (talk) 13:25, 19 February 2016 (UTC)Reply

Hi Alexmar983! Obviously, I like the idea to recruit people to help fill in images on Wikidata (presumably with WD-FIST?). I am a little cautious about mass-spamming, though; maybe a note on the local "village pump" would be less aggressive? Also, be aware that Wikidata currently adds images at ~5x the rate of any other Wikimedia project (see my blog entry), and it has surpassed all but en.wp in total number of items/articles with images. --Magnus Manske (talk) 14:19, 19 February 2016 (UTC)Reply
I have to study WD-FIST, I don't know anything about it. I am leaving for a long flight, so well talk again in some weeks.
In any case I believe the rate is huge but I cannot estimate how many months it will require to get an operational equilibrium. For countries like Italy it's probably a lot, we are talking about years. In their innocence newbies don't lies :D On the bright side, they don't lie when they say they like your tool and they would use it if it were complete, that's why I want to improve the "coverage" of items with images.
Just remember that the local village pump is in theory that on wikidata. We are using the wikimtrics from DEwiki,but we are not leaving a message on DEwiki. But if you prefer we can start with DEwiki village pump, than inform wikidata village pump at a second step, I was hoping to find more de-N interest on wikidata so they could inform later DEwiki because my German is good but not as great as my French or English.
Also, the main target are autopatrolled-like users, so it is not a big "mass". We are talking about a subset of users with a minimum activity on wikidata and editors on dewiki, AND recently active, so it is not a big number. It is spam, but very targeted spam. There are at least two cases of "mass" yet targeted spamming I've co-organized on itWikipedia, and there was no problem. One was about the new CX tool and the ther one about a DRDI information campaign. AP users are traditionally very selective. A 10-33% of them are interested and are grateful, the rest are not interested and skip the message but they don't feel harrassed, I don' t expect DEwiki users to act dfferently, especially when contacted on a different platform. There were even bigger spamming campaigns about WMF election on itWiki and they were not criticized. Recent targeted research has been performed by email from FRwiki, I don't remeber any complaints... I think that past experiences show us that spreading useful information is welcome, a simple discussion is probably enough to take care of all the suggestions. If you want it to be on DEwiki, ok. I preferred data so I could write in English where I master a more technical jargon, but it will just take me 1-2 days more to refine the presntation message.--Alexmar983 (talk) 12:52, 20 February 2016 (UTC)Reply

I have started to spam also WD-FIST, I show it to all users who have started to show interest to the P18 property. It seems ok. I've contacted an user expert on wikimetrics by email, we'll crunch some numbers in the following days. Maybe if you prefer we can target the user active on commons but not active on wikidata. This will also boost the speed of image upload.--Alexmar983 (talk) 01:20, 22 February 2016 (UTC)Reply

wdfist improvement edit

Hi, it's me again. I was playing with the wdfist tool and there is one "option" I cannot find or set up after few minutes of search, so it is possible that some users with less cross wiki experience than me will fac this problem as well. I am preparing a new message for another two local users and I thought I can introduce them to this tool as well, so I wonder if it can be improved. Maybe just a simple shortcut.

So, what "option" are you talking about?

Now my issue is that the tool is good, very handy but suggestions need to be skimmed. That is, too much football-related items for example. Very local users specialized in description of their own territory don't need them IMHO. Is it possible in the wdfist tool to remove all wikidata items without geocoordinates focusing on buildings or places?--Alexmar983 (talk) 14:40, 20 February 2016 (UTC)Reply

Both WDQ and SPARQL queries can filter out without geocoordinates. Property is P625. --2001:630:206:FFFF:0:0:3128:B 15:57, 22 February 2016 (UTC)Reply
The "option" is what is expressed in the secondo part of my comment, that is the possibility to list only items with geocoordinates. I am sure you can get it somehow but can you insert in the list of Permalinks at the end a specific box "show items with geocoordinates (P625)", basically the opposite of what you do with P18. It is much easier for a newbie if they can fill a box, IMHO. I am introducing the tool with the category option, not SPARQL, because SPARQL it is too technical.--Alexmar983 (talk) 18:33, 22 February 2016 (UTC)Reply

Oauth edit


I've been working for a while to help my application shed TUSC so that you no longer have to maintain it. I've proposed an OAuth consumer. Could you authorize it, deny it, let me know what other information you need, etc.?

Thanks. Magog the Ogre (talk) 21:30, 12 March 2016 (UTC)Reply

OK, I think I approved it... --Magnus Manske (talk) 23:45, 12 March 2016 (UTC)Reply

Open Call for Individual Engagement Grants edit


Greetings! The Individual Engagement Grants (IEG) program is accepting proposals until April 12th to fund new tools, research, outreach efforts, and other experiments that enhance the work of Wikimedia volunteers. Whether you need a small or large amount of funds (up to $30,000 USD), IEGs can support you and your team’s project development time in addition to project expenses such as materials, travel, and rental space.

With thanks, I JethroBT (WMF), Community Resources 15:56, 31 March 2016 (UTC)Reply

Thank you edit

Thank you very much for corrections to Bruker:pmt/Bygder -- 16:11, 18 April 2016 (UTC)Reply

Funny F2C result edit


ListeriaBot at iawiki edit

iawiki is maybe the first Wikipedia that has infoboxes on all pages in the article namespace and has no local data in any article infobox - all info there comes from Wikidata.

The next step on automatisation is to have the lists populated from Wikidata. All lists are in namespace 102. [5]

ia:Appendice:Lista de statos de Brasil has Listeria-templates. How can it be started that the bot runs there? 10:31, 20 May 2016 (UTC)Reply

It has picked up automatically by now. Check the bot status if edits seem to lag behind. --Magnus Manske (talk) 12:12, 25 May 2016 (UTC)Reply

Observation regarding geograph2commons edit

I have tried to upload some files from to Commons. Some worked, others not. It appears that only those files could be successfully uploaded on whose pages (not only in the title!) there were no special characters (umlauts, eszett, accented letters) that are not normally used in English. For instance, I never managed to upload pictures from Baden-Württemberg or Köln, or anything with "-straße", "-café" or "-gebäude" in the title or the description. Is this perhaps the reason for the observed inconsistencies? Besides, the script, as it is, still inserts "" in the comment section even if another server is used as the source. --Schlosser67 (talk) 10:30, 10 June 2016 (UTC)Reply

ListeriaBot at eowiki edit

Hi Magnus! I am administrator of Wikipedia in Esperanto. Please, ask for bot status in eowiki for ListeriaBot. Also, global bots are allowed in eowiki. Gamliel Fishkin 11:38, 22 June 2016 (UTC)Reply

Hi, I do not have the bandwidth to ask for bot status on (potentially) 300+ languages, which I do not speak. I applied for a global bot flag, but was told ListeriaBot does not meet the criteria (only specific purposes allowed). --Magnus Manske (talk) 11:48, 22 June 2016 (UTC)Reply
If you want, I can ask for a bot flag for ListeriaBot at Wikipedia in Esperanto (however, AFAIK, in many language sections of Wikipedia an ask to the bureaucrats can be written in English; also, all existing for today bureaucrats of Wikipedia in Esperanto speak German). And, only as a temporary workaround, I can give to ListeriaBot an autopatrolled (autoreview) flag. Gamliel Fishkin 11:35, 26 June 2016 (UTC)Reply

Your Bot has removed Siv-erv_logo1.png from my BNR edit

Hi Magnus Manske, your Bot has removed "Siv-erv_logo1.png" from my BNR entry de:Benutzer:Denalos/Software_Industrieverband_Elektronischer_Rechtsverkehr_SIV-ERV I am a member of this "Industrieverband" and I got the permission to use the logo. Please let me know what's wrong AND until now the article wasn't put into the ANR because it's unter construction. I think it's a bad idea to edit/destroy BNR-Entries without contacting the autor. Best regards --Denalos (talk) 07:33, 28 June 2016 (UTC)Reply

My bot only removes "dead" images, that is, links to images that were deleted on Wikimedia Commons. Deleted file page is here, it was deleted by Jcb. --Magnus Manske (talk) 08:36, 28 June 2016 (UTC)Reply

ListeriaBot@cswiki edit

Hi, do you wish a bot flag for ListeriaBot@cswiki? --Martin Urbanec (talk) 19:12, 8 July 2016 (UTC)Reply

If cswiki wants that, sure. But I can't go through bot applications on dozens of wikis, in languages I don't speak, myself. Only so much bandwidth. --Magnus Manske (talk) 13:13, 9 July 2016 (UTC)Reply
Granted. Thanks for your service. --Martin Urbanec (talk) 13:20, 9 July 2016 (UTC)Reply

OAuth link on CommonsHelper edit


It seems the link to OAuth on CommonsHelper has been removed; I am not sure if this is intentional or not but just wanted to make sure you were aware. MB298 (talk) 19:58, 17 August 2016 (UTC)Reply

Thanks, fixed! --Magnus Manske (talk) 20:08, 17 August 2016 (UTC)Reply

LdiF as Mix'n'Match edit

Hey Magnus, how can I add {{{}}} to Mix'n'Match? Is it enough to give a CSV-File with the titles and ids or sth. like that? Queryzo (talk) 20:47, 8 September 2016 (UTC) Queryzo (talk) 20:47, 8 September 2016 (UTC)Reply

Yes, just paste or upload a tab-delimited file (like, "copy Excel into Notepad") here. --Magnus Manske (talk) 19:27, 10 September 2016 (UTC)Reply

Funktioniert super! Könntest du die erste Version löschen (oder verstecken) [6] Bei der zweiten Version bitte den Zusatz " (new)" aus dem Titel raus. [7] Danke. Queryzo (talk) 06:06, 4 October 2016 (UTC)Reply

Erledigt. --Magnus Manske (talk) 08:41, 4 October 2016 (UTC)Reply

Danke! Die auto matches sind wirklich gut. Ich finde kaum einen Fehler. Wäre es möglich, alle auto matches für Wikidata-Items vom Typ Film (+ Unterklassen) automatisch zu bestätigen? Sofern Filme gleich benannt sind, kriegen sie im Lexikon wie in der Wikipedia einen Klammerzusatz. Queryzo (talk) 19:42, 5 October 2016 (UTC)Reply

Würde ich lieber nicht machen. Namen-matches können sehr gut sein, aber der Punkt von mix-n-match ist die manuelle Kontrolle. Ich mache manchmal Ausnahmen, aber dann z.B. matching über externe IDs. --Magnus Manske (talk) 08:57, 6 October 2016 (UTC)Reply

ListeriaBot edit

Looks like ListeriaBot does unneeded page edits -- it only changes order of rows, but doesn't add data: --WikedKentaur (talk) 10:23, 18 September 2016 (UTC)Reply

Mix'n'Match OFDb edit

Hallo Magnus, ich bin bereit für den zweiten Film-Katalog. Können wir noch was verbessern, z.B. Film als type, damit nicht die ganzen Personen und Romane Auto match verwirren ? Queryzo (talk) 16:46, 11 October 2016 (UTC)Reply

"Type" unterscheidet im Moment nur zwischen "person" und "sonstiges". Keine Spezialbehandlung für "film" etc. --Magnus Manske (talk) 13:50, 13 October 2016 (UTC)Reply
eben! Wäre es aufwendig, diese nachzurüsten? Queryzo (talk) 20:31, 13 October 2016 (UTC)Reply
Erledigt. Gib als Typ die Q-ID des "Wurzelelements" an (hier wohl Q11424 für "Film"), damit auch Unterklassen (z.B. "Dokumentarfilm") berücksichtigt werden. Habe das mal für LdiF-IDs gemacht; auto-matches sind von ~21% auf ~17% zurückgegangen, aber dafür hoffentlich mit höherer Qualität. --Magnus Manske (talk) 09:14, 14 October 2016 (UTC)Reply
Super, danke! Queryzo (talk) 17:07, 14 October 2016 (UTC)Reply

Der neue Katalog ist nun online: [8] Allerdings hat sich trotz Angabe von Q11424 einiges an Nicht-Film-Items angefunden. Kannst du mal gucken? LG Queryzo (talk) 17:57, 19 October 2016 (UTC)Reply

Das kommt daher, das in Deinem Import Schrott-Typen sind, z.B. "bestia magnifica", und nicht nur Q11424. Vergessen, Tabs rauszufiltern? --Magnus Manske (talk) 18:43, 19 October 2016 (UTC)Reply
Update: Habe die Typen repariert und neu laufen lassen. --Magnus Manske (talk) 19:25, 19 October 2016 (UTC)Reply
Ah, hab mich schon gefragt, was die kryptischen Typen bedeuten. Ich nehme mal, dass die Tabs in den Titeln drin waren und der Typ dann als 4. Spalte einging? Wäre dann wichtig zu wissen für den nächsten Katalog, wie man das verhindern kann. Queryzo (talk) 19:59, 19 October 2016 (UTC)Reply

Flickr2commons edit

Flickr2commons stopped working. Please have a look. Thanks. Materialscientist (talk) 10:37, 21 October 2016 (UTC)Reply

Try it now. --Magnus Manske (talk) 17:05, 21 October 2016 (UTC)Reply
Tried with two different images (same uploader though), on two different PC platforms - no luck. Materialscientist (talk) 20:56, 21 October 2016 (UTC)Reply
Works now (something has been reset, as I was asked to re-authorize flickr2commons). Materialscientist (talk) 05:44, 22 October 2016 (UTC)Reply

Templator formatting edit

Hey Magnus, could you please check out the Templator (Vorlagen-Generator) DE and the reference "Internetquelle"? The column in the middle where all the data goes is really small. It's much large in EN version. Is it possible to fix this? Also could you please have a look at the change request over here? A space is missing between the parameter and the |. Thanks in advance it's a great tool, I use it daily. --Abu-Dun (talk) 08:48, 15 November 2016 (UTC)Reply

Oh and would it be possible to make it possible to link to weblinks in the "Explanation and example" ("Beispiel/Kommentar") column? It would be great to get further information to some keywords like ISO 639. Right now the links aren't properly formatted, see Vorlagen-Generator: Internetquelle --Abu-Dun (talk) 08:54, 15 November 2016 (UTC)Reply

Neue Kataloge edit

Hallo Magnus, wieder zwei neue Kataloge: 293, 294. Hier bitte wieder die Typen korrigieren (293 = "person", 294 = Q11424). Zudem ist mir beim 294 nun doch der Titel "Filmportal Film-IDs" lieber. Daaaanke! Queryzo (talk) 20:54, 16 November 2016 (UTC)Reply

Erledigt. --Magnus Manske (talk) 22:11, 16 November 2016 (UTC)Reply
Danke. Kannst du bei 294 noch einen neuen Suchdurchlauf für die Automatches machen? Queryzo (talk) 06:38, 17 November 2016 (UTC)Reply
Läuft. --Magnus Manske (talk) 09:00, 17 November 2016 (UTC)Reply

Korrektur in Mix'n'match bei Merge edit

Hallo Magnus, könnte man die Verknüpfung in Mix'n'Match automatisch korrigieren, wenn ein Item gemergt wird? Derzeit ist das mühsame Handarbeit über "Sync", Quick Statements und die Suchfunktion in Mix'n'match. Queryzo (talk) 06:53, 14 December 2016 (UTC)Reply

Das wurde sowieso schon automatisch gemacht, aber immer nur für einen zufälligen Teil von mix'n'match. Habe noch was gebaut, das jetzt jeden Tag einmal "hinterherräumt". --Magnus Manske (talk) 23:54, 14 December 2016 (UTC)Reply

Feature gone edit

Hi Magnus Manske, I previously used the Autolist tool to quickly download from a category the Q IDs of articles, and this option seemed to have gone in Petscan. Can the option to download a the results be re-implemented. Thanks! Romaine (talk) 15:37, 30 December 2016 (UTC)Reply

File:Citation screenshot Kopie.png edit

Hello Magnus. Happy new year. I've found an old file by you at File:Citation screenshot Kopie.png and I had to tag it with GFDL-presumed. If you can clarify the status of the copyright on the image it'd be awesome. I'm trying to review all the files hosted here and move the appropriates to Commons. Best regards, —MarcoAurelio 16:37, 3 January 2017 (UTC)Reply

Mix n' Match request edit

Internet wrestling database

Type "Episodes" in Mix'n'Match edit

Hallo Magnus, gestern habe ich wieder ein paar Kataloge mit Episoden eingestellt. Ich habe gesehen, dass man nach Typ sortieren kann, evt. wäre es auch sinnvoll, den Typ "Episodes" o.ä. einzuführen, dann hätte man die gebündelt. Werden ja wahrscheinlich noch einige dazukommen. Queryzo (talk) 07:35, 17 April 2017 (UTC)Reply

And that seems to be not correct. Queryzo (talk) 11:18, 22 April 2017 (UTC)Reply

Mix'n'match - same name edit


is it possible to bring back the same name feature in Mix'n'match? It was really useful and a quick way to add matches. Thanks! --Adam Harangozó (talk) 22:40, 4 May 2017 (UTC)Reply

Would Creation candidates serve as a substitute, or is there anything specific about "same name" you are missing? --Magnus Manske (talk) 08:36, 11 May 2017 (UTC)Reply

V&A Museum items edit

Can you create a catalog of items from to correspond with P3929

Done. --Magnus Manske (talk) 14:01, 13 May 2017 (UTC)Reply

Prelinger Archives edit

Can you add the active film archive list to Mix n Match? --MistressData (talk) 03:35, 17 May 2017 (UTC)Reply

Done. --Magnus Manske (talk) 13:26, 18 May 2017 (UTC)Reply

Board Game Geek Games edit

Could you make a catalog based on to correspond with P2339--MistressData (talk) 19:00, 17 May 2017 (UTC)Reply

Done. --Magnus Manske (talk) 07:34, 19 May 2017 (UTC)Reply

Arcade History edit

Could you make a catalog based on this site? --MistressData (talk) 20:00, 22 May 2017 (UTC)Reply

Done. --Magnus Manske (talk) 10:26, 23 May 2017 (UTC)Reply

Mix'n'Match ČSFD edit

Hello, I tried to import part of movie database and found some issues:

  • Because of no API on the CSFD I made some list from source html and got 300 entries. Is possible to import new entries to this catalog? What if there will be duplicates?
  • Many items from this catalaogue is already matched with WD entries (minimally 250/300), but automatching gives only 150 results and on the rest sometimes is the correct possibility via SPARQL, but sometimes not even if the WD entry already have CSFD ID filled. See [9] vs. [10].

JAn Dudík (talk) 07:55, 21 June 2017 (UTC)Reply

I am trying to import the entire thing now. I believe I can have it auto-update as well. Initial import may take a day or so. --Magnus Manske (talk) 08:35, 21 June 2017 (UTC)Reply
OK, 71587 imported. Didn't get any more (unique) hits from the website, though it seems there are more. --Magnus Manske (talk) 07:24, 22 June 2017 (UTC)Reply
Great! There is another big set - people (P2605), example (WD), can you import this too? JAn Dudík (talk) 07:37, 22 June 2017 (UTC)Reply
Is there a list or search function for people on that site, like there is for movies? Can't see it, and I don't speak cs --Magnus Manske (talk) 07:48, 22 June 2017 (UTC)Reply JAn Dudík (talk) 19:02, 22 June 2017 (UTC)Reply

Mix'n'match importing edit

I've been playing around with the Mix'n'match tool (which is amazing, thank you!) and want to try importing some catalogs. Is there a way to import catalogs which do not have unique numerical identifiers? In particular I'm thinking of some internet and print reference works, like The Encyclopedia of Science Fiction or the Encyclopedia of Library History. Thanks. Gamaliel (talk) 23:31, 29 June 2017 (UTC)Reply

Mix n match aranea edit

Hi, can you do (P3594) Mix n match catalog. Thanks

Done. --Magnus Manske (talk) 07:00, 21 September 2017 (UTC)Reply

Mix'n'match TGN URLs? edit

Hello! I don't know if you respond to pings on other pages, but could the URL which entries in catalogue 49 (TGN IDs) use be changed to "$1"? This URL, as opposed to the one ("$1") which is used at the moment, actually displays synonyms for the given place name (indicating its level of administrative division) and the appropriate position in the geographic hierarchy of the place. To find these with the existing URL requires two more clicks to get to the URL I'm suggesting, which slows down identification of the place. Mahir256 (talk) 14:10, 22 September 2017 (UTC)Reply

Done. --Magnus Manske (talk) 15:09, 22 September 2017 (UTC)Reply

Mix'n'match importer example edit

Hi, can you make example of what should Mix'n'match importer look like when its completed. Thanks

It should tell your that your catalog has been created, with a link. Import should take <1min, unless it's especially large (>50000 entries or so). --Magnus Manske (talk) 18:02, 13 October 2017 (UTC) ¨Reply
I mean example of creating one. --Walter Klosse (talk) 19:00, 13 October 2017 (UTC)Reply
Like this? (Gerne auch auf Deutsch!) --Magnus Manske (talk) 11:47, 14 October 2017 (UTC)Reply
Thanks! btw.: ich spreche nicht gut auf Deutsch --Walter Klosse (talk) 17:46, 18 October 2017 (UTC)Reply
What should i do when i mistakenly upload only 5 items (Link)? --Walter Klosse (talk) 17:51, 18 October 2017 (UTC)Reply
sigh email me the file so I can do it myself... FIRSTNAMElastname(at)gmail com --Magnus Manske (talk) 22:27, 18 October 2017 (UTC)Reply
Here you got link, but i will email the file too. --Walter Klosse (talk) 12:28, 21 October 2017 (UTC)Reply
Done. --Magnus Manske (talk) 13:08, 22 October 2017 (UTC)Reply

Catholic Encyclopedia ID edit

Mix'n'match says the Catholic Encyclopedia catalog does not have a Wikidata property. I guess it's because it's one of the first sets, but P3241 has been created a while ago. I assume only you can update the catalog with the Wikidata property association. Could you do that? Ijon (talk) 20:01, 24 October 2017 (UTC)Reply

That is indeed one of the reasons. The other is that it was taken from Wikisource rather than from newadvent. I have created a new catalog for this, because the original does not use the same IDs. I will try moving matches over where I can automatically based on name, but I would like to close the old catalog eventually, even if we lose some matches, and start afresh. What do you think? --Magnus Manske (talk) 13:58, 25 October 2017 (UTC)Reply
Sure, let's start afresh. Ijon (talk) 14:59, 25 October 2017 (UTC)Reply
OK, I have imported (as much as I could) the old matches into the new catalog. The old one is now deactivated. I am running automated matching on it now, but it's as good as it gets for now. Syncing with Wikidata is under way. --Magnus Manske (talk) 15:32, 25 October 2017 (UTC)Reply

Bug of ListeriaBot? edit

Dear Magnus,
I wanted to take a moment to post a message about Listeriabot activities on the French-speaking WP. I have seen here that it removed quite a lot of links to Wikidata for articles that do not exist in French yet and, as far as I can see, there has been no change to the query definition nor to the Wikidata elements that trigger links to appear in this list (ex. Q4612975 has been removed but nothing has been changed recently). Can you try to have a look to understand the root cause? Regards, Moumou82 (talk) 18:26, 4 January 2018 (UTC)Reply

The query on that page gets the first 1000 Wikidata items related to Tunisia (country:Tunisia) with no French page associated. But, the first 1000 of what? The order is not defined in any way, and made up by the query service. So if the order changes, you get different 1000 items. I suspect that the latest items with country:Tunisia are put first; for example, this edit happened just hours before the update.
This query should be a bit more stable, as it defined an order (by item). It also gets you an actual 1000 items through the use of DISTINCT. --Magnus Manske (talk) 10:45, 5 January 2018 (UTC)Reply
Many thanks Magnus! Moumou82 (talk) 19:11, 5 January 2018 (UTC)Reply

Source code for Reasonator? edit

Hi! I wanted to try out the source code of Reasonator by myself and see how it works, I downloaded it from here: But I get this message: "Warning: require_once(/data/project/reasonator/public_html/php/common.php): failed to open stream: No such file or directory in /srv/disk1/2462897/www/ on line 3. It appears that I must have a file called common.php, but it doesn't exist in your source code. How do I do? /user:bollkalle_55 2018-04-22 23:19 (Swedish time)

You'll need some more code from here. --Magnus Manske (talk) 08:23, 23 April 2018 (UTC)Reply

GNIS Catalog Property Corrections from P804 to P590 - oops edit

Please correct the Property for the following GNIS catalogs in Mix'n'Match, I accidentally swapped properties in my head when filling out the fields. First import was correct (GNIS Lakes), and I failed for the remainder. I didn't want to import any more until this is corrected. I'm being more careful now.

I changed all of these properties to P590. Also did 1126, as it was using P804 too, and was created by you. Let me know if I should change it back. --Magnus Manske (talk) 07:45, 2 May 2018 (UTC)Reply
Thanks for catching 1126. I guess the last task to clean this up is correcting the manual and auto-matches that occured before the property correction that still use P804 instead of P590. If you have a script to tackle this cool, else I'll tackle this. --Wolfgang8741 (talk) 01:58, 3 May 2018 (UTC)Reply
I think I've cleaned up the previous auto-matches and manual matches. There weren't too many to do. Thanks. --Wolfgang8741 (talk) 03:45, 3 May 2018 (UTC)Reply

LoC relators Mix'N'Match catalogue (1277) - ID correction? edit

I created the LoC relators catalogue for the P4801 property--my first catalogue--but made a mistake by providing just the relator ID for each entry (e.g. "abc") instead of the fuller path necessary for the LoC vocabularies ("relators/abc").

Would it be possible to update the catalogue to prepend "relators/" to all of the IDs? Dan scott (talk) 19:05, 5 June 2018 (UTC)Reply

Done. --Magnus Manske (talk) 22:18, 5 June 2018 (UTC)Reply

Artikel "Günter Jendrich" edit

Lieber Herr Manske. JCB hat das Foto "Günter Jendrich - Thorsten Jendrich.jpg" aus dem Artikel "Günter Jendrich" in der dt. Wikipedia entfernt mit dem Argument, es läge keine Genehmigung für die Veröffentlichung des Bildes vor. Das ist aber nicht der Fall! Der Fotograf und Eigentümer des Bildes, Herr Throsten Jendrich (er ist der Sohn von Günter Jendrich und hat das Foto 1968 auf einer privaten Feier gemacht) hatte die Erlaubnis zur Veröffentlichung unter den Wikipedia-Bedingungen sowohl an als auch an geschickt. Von daher bitte ich darum, das Bild wieder in den Artikel einzupflegen. Beste Grüße Josefblau

File:Newlogo2.png, File:Newlogo2a.png, File:Newlogo2b.png edit

Used at Logo_suggestions#2,_3,_4, any comment for Meta:Requests for deletion, please?--Jusjih (talk) 21:04, 4 August 2018 (UTC)Reply

User:CommonsDelinker edit

Dear Magnus, please look there: Yours sincerely --Sockenschütze (talk) 19:20, 22 August 2018 (UTC)Reply

Import Mix'n'match catalogues edit

Can you help me to import Mix'n'match catalogues? Ilham151096 (talk) 21:55, 4 September 2018 (UTC)Reply

I can try. What specifically? --Magnus Manske (talk) 09:20, 5 September 2018 (UTC)Reply

Search from Mix'n'match often not working edit

Hi Magnus, using the visual tool, often the wikidata search is not working and gets stuck at "Searching...". Here are some example queries that did not work:

Chang and Eng

John Kasich

L H Myers

Haakon II Sigurdsson

Thomas Hartman

FS100 (talk) 12:25, 19 September 2018 (UTC)Reply

I tried "John Kasich", which works for me. Mix'n'match, as a whole, sometimes gets stuck as a result of catalog-combining searches (not from the visual tool), and because the MySQL version on Toolforge is too old to use per-query runtime limitations. I was told this will be fixed at some point. --Magnus Manske (talk) 21:29, 19 September 2018 (UTC)Reply

Mix'n'Match catalog request edit

Hey Magnus! How are you? I hope everything is fine; Brazilian Wikimedians send regards :)

Could you please create a catalog for Mix'n'Match on d:Property:P5950? I might have another similar request pretty shortly (d:Wikidata:Property proposal/Museu Nacional ID) Thanks much! --Joalpe (talk) 21:59, 5 October 2018 (UTC)Reply

Now building here. --Magnus Manske (talk) 13:46, 6 October 2018 (UTC)Reply

New Mix'n'Match scraper fix? edit

Hi Magnus, I already asked this on Twitter but then figured here was probably a better place. I created a new scraper but messed up the URL part when I saved it (basically put "/(.+?).html" instead of "/$1.html"). I corrected it and saved again (with the new catalog ID), but the links still all go to "/(.+?).html". Like, literally. Can it be fixed? Or at least completely deleted so I can start over and do it right this time? Thanks! ---LadiesMakingComics (talk) 21:05, 22 October 2018 (UTC)Reply

Oh boy, I couldn't leave well enough alone, could I? So I also made another new catalog and this time I messed up the RegEx to swap the names around, so most were getting pulled in as "Lastname, Firstname". While not ideal, I wasn't going to try to fix it until I realized that it might be impeding automatic matches (and with 32k entries, that was a problem). So I added a "new" scraper just to fix that issue. I figured that if nothing else, it would apply to entries added to the site later. But what it actually did was pull all the names again. HELP!!!! --LadiesMakingComics (talk) 20:15, 23 October 2018 (UTC)Reply

British Museum Thesaurus not working edit

Hey, the [11] catalog's links seem dead because they changed the link structure. Unfortunately I don't know how, could you repair the scraper and the wikidata property? Thank you! --Adam Harangozó (talk) 12:32, 29 October 2018 (UTC)Reply

Also, is it possible to edit existing scrapers to add catalog type where it's set as unknown? (The catalog editor option doesn't let me change it.) --Adam Harangozó (talk) 15:28, 30 October 2018 (UTC)Reply
I also brought this up on the talk page of the BMT ID property and it unfortunately goes beyond just the link structure, they also changed the ID numbers. --Lexid523 (talk) 19:53, 9 November 2018 (UTC)Reply

Emory Women Writers Resource Project edit

Hi Magnus, any chance you can make a scraper of authors from the Emory Women Writers Resource Project? Whenever you have the time. Thanks!! --Lexid523 (talk) 15:37, 5 November 2018 (UTC)Reply

Done. --Magnus Manske (talk) 16:15, 5 November 2018 (UTC)Reply
Great! Thanks again! --Lexid523 (talk) 16:57, 5 November 2018 (UTC)Reply

Orlando Project edit

So, now that that's taken care of, there's another women writers' project database I'd like to import, but its set-up is pretty complicated (to me at least) and I have no hope of figuring it out on my own. I've gotten as far as figuring out that THIS is the link that will get you to the list of names with all the correct parameters. The links in that list all lead to similarly constructed URLs, like this:

<a href="../protected/svPeople?formname=r&subform=1&person_id=[6-LETTER AUTHOR ID]&adt_start_year=0612&crumbtrail=on&dt_end_cal=AD&dt_end_day=05&dt_end_month=11&adt_end_year=2018&dt_start_cal=BC&dts_historical=0612--+BC%3A2018-10-19&dts_lives=0612--+BC%3A2018-10-19&dts_monarchs=0612--+BC%3A2018-10-19&ls_bww=on&ls_iww=on&submit_type=A-Z">[Lastname, Firstname]</a>

So if you can work with that, or can figure out a better back door way, I would greatly appreciate it. But again, no rush! Thanks! --Lexid523 (talk) 21:16, 5 November 2018 (UTC)Reply

Done, but queued (Toolforge is having a bad day again, just keep reloading...). FWIW, I used the regex
<a href="../protected/svPeople\?.*?person_id=([a-z]+).*?>(.+?)</a>
Note that the Mix'n'match scraper doesn't work with IDs that have trailing spaces, so some URLs, like for "aria", don't work. --Magnus Manske (talk) 09:27, 6 November 2018 (UTC)Reply
You are the best! --Lexid523 (talk) 16:26, 6 November 2018 (UTC)Reply

Mix'n'match Scraper for edit

Hi Magnus! I have been trying to add a scraper for (P480) without success. I am just getting an error while trying to test it (ERROR: Test has failed, for reasons unknown. Could be the server to be scraped is too slow?). The following are the parameters I am using, I am hoping you can get it working (or advice me how to fix).

Catalog name:
Description: Movies database.
WD property: P480
Type: Media
Primary language: en
Level 1 keys: 0-9, A...Z (note, "0-9" as is, not a range)
Level 2 range: Start 1, End 2000, Step 1 (I put 2000 because the key T goes up to 1627 (here)
URL Pattern:$1_$2.html
RegEx entry: <a href="\/en\/film(\d{6})\.html" title="(.*)">(.*)<\/a> \((\d{4})\) (I tested it on and it is working. The output will be ID, movie name (twice) and then year)
id: $1
name: $2
desc: $4 (I am just throwing the year in the description)
url: https:///$1.html
type: Q11424

I really have no idea what is the issue. Thanks in advance and keep up the good work!
--Cabeza2000 (talk) 23:20, 30 November 2018 (UTC)Reply

You don't need and shouldn't protect the /, and avoid the .*
This work : <a href="/en/film(\d{6}).html" title="([^"]+?)">([^<]+?)</a> \((\d{4})\).
Compress whitespace must be checked. Eru (talk) 21:57, 2 December 2018 (UTC)Reply
Thanks Eru, it worked. I saved it as catalog 2048, it will probably take a while to run.--Cabeza2000 (talk) 00:09, 3 December 2018 (UTC)Reply

PetScan issue edit

Please have a look at the problem. Thanks! --Red Winged Duck (talk) 12:52, 11 December 2018 (UTC)Reply

BaGLAMa 2 request edit

Hello Magnus, would it be possible to add the Commons categories Images from the Royal Museum of Fine Arts Antwerp and Images from the King Baudouin Foundation to the BaGLAMa 2 list? Is this also possible retroactively, starting from March 2018? Many thanks! Beireke1 (talk) 12:05, 4 January 2019 (UTC)Reply

Both added now. I don't look at meta often, Wikidata is better to leave requests! --Magnus Manske (talk) 14:07, 27 March 2019 (UTC)Reply

Wikidata todo edit

Wikidata to do, specifically duplicity from Wikispecies, in inaccessable for 2 days. Also can't get on your bug report page. What happened? Neferkheperre (talk) 18:30, 17 February 2019 (UTC)Reply

@Neferkheperre: there are issues with the WMFLABS servers that have spread to Tools. Some things are being moved, and will do so gracefully, some will need intervention. Patience is required whilst WMF fixes the servers and recovery occurs.  — billinghurst sDrewth 22:54, 17 February 2019 (UTC)Reply
Thanks. Let me know. I have been creating Wikidata pages for Wikispecies articles for a couple of months now, using the duplicity function. Neferkheperre (talk) 23:23, 17 February 2019 (UTC)Reply

Scraper request edit

Hi, could you make a scraper for the periodicals in the Hungarian Electronic Periodicals Archive & Database ([12])? Thanks! - Adam Harangozó (talk) 13:49, 27 March 2019 (UTC)Reply

Should be scraping now, here. I don't look at meta often, please leave future requests on my Wikidata talk page, thanks! --Magnus Manske (talk) 14:06, 27 March 2019 (UTC)Reply

ERROR: Less than three coulmns on row 1 edit

Hi Magnus, I'm trying to import a catalog but keep getting "ERROR: Less than three coulmns on row 1" when trying to upload a .csv file with postal codes. Here's a sample of what they look like.

60-0000 以下に掲載がない場合 Ikanikeisaiganaibaai, Sapporo Shi Chuo Ku, Hokkaido 60-0000
64-0941 旭ケ丘 Asahigaoka, Sapporo Shi Chuo Ku, Hokkaido 64-0941
60-0041 大通東 Odorihigashi, Sapporo Shi Chuo Ku, Hokkaido 60-0041

The fields are ID, name, description and URL (not really a URL, I just want to match localities with ZIP codes). Any idea what I could be doing wrong? NMaia (talk) 07:26, 3 April 2019 (UTC)Reply

You say CSV file, but your example is tab-delimited, which is good. Maybe you have a header line in there? --Magnus Manske (talk) 15:15, 12 April 2019 (UTC)Reply
I don't think so. Here's the file: NMaia (talk) 14:07, 29 April 2019 (UTC)Reply
OK, I downloaded the file. First, it needs to be TAB-separated, not COMMA-separated. Second, no quotes. Third, there is a fourth column that seems to repeat the ID from the first one. Fourth, there are duplicate IDs (e.g. "4-0000") which doesn't work in Mix'n'match. --Magnus Manske (talk) 15:01, 29 April 2019 (UTC)Reply
Thanks! I'm looking into ways of cleaning up this dataset. Is there anything that can be done about these IDs? They are postal codes, and, although the same, seem to refer to different places in Japan. NMaia (talk) 06:40, 30 April 2019 (UTC)Reply

Mix'n'Match Catalog for PRELIB Collectifs edit

Hello, I am trying to create a new Mix'n'Match catalog to scrape collectifs on PRELIB and to link the collectifs to But I have some troubles with the method. Could you give me a hand ? I tried the following:

URL Follow level:$1
regex Follow level: \d+

URL pattern:$1
RegEx entry: <a href="/prelib/collectif/(\d+)/">(.+)</a>

id: $1
name: $2

But when I test the scraper, the following error is displayed: ERROR: Test has failed, for reasons unknown. Could be the server to be scraped is too slow?

Thanks, Hopala! (talk) 08:02, 1 May 2019 (UTC)Reply

I tested another parameters which seem to work:

I used a range level, since I could not figure out how to manage a follow level.

URL pattern:$1
Regex block: <table>(.+)</table>
RegEx entry: <a href="/prelib/collectif/(\d+)/">([^<]+?)</a>

Hopala! (talk) 10:25, 1 May 2019 (UTC)Reply

mix n match help, AAAS fellows edit

Hi! I'm working on a project to upload all the engineering fellows (current and historic) in the American Association for the Advancement of Science into Wikidata, as they are all notable enough to have Wikipedia biographies. The current fellows listing is here but only Section "Engineering (Section M) lists the engineering fellows. I'm not sure how to upload them, but I did make a file for them, but I'm not sure the entry ID is correct. Mix n Match is telling me I didn't list the url or url code, but they don't have them. What should I do? Thank you, and thank you for the time you took making this tool! It's super cool Sbbarker19 (talk) 17:18, 7 May 2019 (UTC)Reply

I made a scraper for all (not just engineering) here, will take a few hours to fill/show. --Magnus Manske (talk) 12:46, 8 May 2019 (UTC)Reply

Import large file / new version NKČR to MixNMatch edit

Hi Magnus, ill preparing large version of file (and with better structure to NKČR catalog - I uploaded 100000 records and its about 10% of whole catalog. And unfortunately with bad "name". Is there possibility to upload file other way? And remove currently data. TSV data is - thanks if its possible. --Frettie (talk) 11:01, 26 June 2019 (UTC)Reply

Importing now. --Magnus Manske (talk) 09:06, 27 June 2019 (UTC)Reply
Turns out your file contains 828508 rows but only 300000 unique ones. Did you run into some limit somewhere? --Magnus Manske (talk) 09:43, 27 June 2019 (UTC)Reply

Can't change scraper after creation on mix'n'match edit

Hey there, I created this scraper last night and wanted to change it afterwards because of a mistake I made with the link form the items to the db. And also I wanted to add a description with a new regex. So I went here and tried to entered this:

Catalog ID: 2651

Catalog name: OpenCritic game ID

Description: Identifier for a game on


WD property: P2864

Type: Video games

Primary language: en

One Range Level: 0-100, Step 1

URL pattern:$1

RegEx entry: <a _ngcontent-(.{1,5})="" href="/game/(\d{1,5})/(.{1,100})">(.{1,100})</a></div><div _ngcontent-(.{1,5})="" class="platforms col-auto"> (.{1,50}) </div><div _ngcontent-(.{1,5})="" class="first-release-date col-auto show-year"> (.{1,50}) </div>

id: $2

name: $4

desc: Released: $8 Platform: $6


type: Q7889

And activated "Compress whitespace". Then tested it, with the right result and hit "Save Scraper/Catalog". But it doesn't seem to have updated at all. (I also tried to do it with out entering anything in the first section except the Catalog ID) Should I create a new one or could you help me fix this one? Greetings, and thanks for all the tools --Kristbaum (talk) 10:01, 28 June 2019 (UTC)Reply

Mix 'n' Match Landsholdsdatabasen edit

Would you kindly import {{{}}} into Mix 'n' Match? The complete list of entries can be found here in Danish. --Trade (talk) 23:08, 7 July 2019 (UTC)Reply

I suspect you meant ? Import running. --Magnus Manske (talk) 10:13, 8 July 2019 (UTC)Reply

FIST error running an empty query? edit

Hi, just wanted to say I have filed an issue #10 on FIST at BitBucket

--CiaPan (talk) 10:16, 4 August 2019 (UTC)Reply

Set existing Mix'n'Match scraper to auto-update? edit

Hi Magnus, I put this in the Manual talk a bit ago but realize that might have been fruitless. Is there any way to set an existing scraper to regularly update via automated web page scraping? I made three catalogs with (what I thought was) the exact same process but only LezWatch.TV actors is missing the line that says it will regularly update. Can I change a setting somewhere to have this one auto-update as well? Sweet kate (talk) 19:11, 18 August 2019 (UTC)Reply

I have switched on auto-update for #2726. --Magnus Manske (talk) 08:56, 27 August 2019 (UTC)Reply

BaGLAMa 2 request edit

Hi Magnus, could you perhaps add commons:Category:Media_contributed_by_Biblioteca_Nacional_Aruba to to BaGLAMa 2? Thanks in advance! --BNA-PS (talk) 13:36, 30 August 2019 (UTC)Reply

Done. --Magnus Manske (talk) 13:36, 5 September 2019 (UTC)Reply

Baglama2 request edit

Hello Magnus, Netha Hussain and I are doing a project to upload all 3000 images from to Commons, currently pm 15% done, see c:Category:Media_from_SMART-Servier_Medical_Art. As we would like to track reuse and uptake, could you pls add that category to the Baglama2 tool? Many thanks, --OlafJanssen (talk) 20:36, 30 September 2019 (UTC)Reply

Done. --Magnus Manske (talk) 07:28, 1 October 2019 (UTC)Reply
Super, many thanks!! --OlafJanssen (talk) 14:03, 1 October 2019 (UTC)Reply

Mix'n'match FemBio, remove duplicate edit

I created but it is a duplicate of , so please remove it if you can. Thanks --PeterTheOne (talk) 16:37, 20 October 2019 (UTC)Reply

Deactivated. --Magnus Manske (talk) 09:32, 21 October 2019 (UTC)Reply

Add Auckland Museum Images to BaGLAMa 2 edit

Hi Magnus, would you be able to add Auckland Museum Images to BaGLAMa2 please? We have 100,000+ images on wikimedia commons. Category is "Images_from_Auckland_Museum" Best, James

Thanks for adding Magnus! Cheers, James

Gelöschte Bilder aus deutschem Artikel und Wikimedia "Stuart Wolfe" edit

Hi Magnus, Dein Name klingt deutsch, daher versuch ich's mal in meiner Muttersprache.

Als Betreuer des CommonsDelinker appelliere ich an Dich, die Löschungen wieder rückgängig zu machen. Zu allen markierten Bildern wurden mithilfe des Rechte-Assistenten alle erforderlichen Informationen nachgereicht. Diese beinhalteten u.a. ein persönlicher E-Mail-Link zum Künstler selbst, Stuart Wolfe, sowie ein Cloud-Link zu einem Selfie, das den Künstler in seinem Atelier zeigt, mit einem Blatt Papier in der Hand, das mich, den Artikel Autor "Smokeonthewater", zum Verwenden der Bilder in Wikipedia ermächtigt und die Liste der markierten Dateien nennt. Offenbar wurden diese Informationen von Deinem Bot oder Dir ignoriert. Gelöscht wurden dabei Fotos, die ich im Atelier des Künstlers persönlich aufgenommen habe. In Anbetracht meines langjährigen Status' in Wikipedia finde ich das einfach nur respektlos. Daher bitte ich um eine konstruktive Antwort sowie einen Vorschlag, wie der Zustand des Artikel möglichst ohne großen Aufwand wiederhergestellt werden kann. Ich habe einschließlich des Besuchs im Künstleratelier etwa 20 Stunden ehrenamtliche Arbeit investiert und bin sehr verärgert über die Situation.

Viele Grüße, --Smokeonthewater (talk)

Hallo Smokeonthewater, Deutsch passt in der Tat! Zur Sache: Der CommonsDelinker löscht keine Bilder, sondern entfernt nur "Bild-Links" zu bereits gelöschten Bildern, die (zur Zeit der Löschung) sowieso nicht dargestellt werden können. Die Entfernung der "Bild-Links" wird zudem von Commons-Admins in Auftrag gegeben. Ich und mein Bot haben weder etwas mit der Löschung der Bilder auf Commons, noch mit der Erteilung des Entfernungs-Auftrags zu tun. Der Edit auf de:Stuart Wolfe wäre natürlich leicht rückgängig zu machen, sähe dann aber so aus, denn das Bild ist auf Commons immer noch gelöscht. Um das Bild wieder herzustellen, sprich am besten mit dem Admin, der es gelöscht hat, oder auf einer der Admin-Diskussions-Seiten. --Magnus Manske (talk) 08:34, 11 December 2019 (UTC)Reply
Hallo Magnus, vielen Dank für die Erläuterungen und den Verweis an den Admin. Sorry, wenn ich Deine Rolle falsch eingeschätzt habe. --Smokeonthewater (talk)

BaGLAMa 2 Request for National Gallery of Art edit

Hi Magnus,

Would you be able to add the Category:Images from the National Gallery of Art to BaGLAM 2? Thanks. -- brwz (talk) 14:05, 18 December 2019 (UTC)Reply

Now here. --Magnus Manske (talk) 15:12, 18 December 2019 (UTC)Reply

Reset Mix'n'match catalog #3216 edit

Hi Magnus, would you be able to delete reset (remove all entries) in ? I have better regex for the name field now but it appears that (1) I cannot remove entries from the existing catalog, and (2) I cannot create a new catalog with the same info (perhaps because the WD property is in use?) [it was because the name was taken, so now I've created a replacement catalog] Also (3) it appears that "Catalog editor" doesn't save any of my changes (e.g., name/title change). Am I doing something wrong? And would you be able to reset #3216 so I can get started with proper matches? Thanks for a great tool! czar 11:40, 28 December 2019 (UTC)Reply

Also I wasn't able to convert the surnames to title case with regex tricks, if that's something you can do on your end. czar 23:21, 28 December 2019 (UTC)Reply
  • One more :) I noticed that when I "confirm" an "Automatically matched" item, the claim is added without a reference, but if I create a "new item" (when a Q doesn't already exist), the claim is added with a reference. czar 00:30, 29 December 2019 (UTC)Reply

Scaling WikiData edit

Hello, thanks for your kind message on the original Grant Proposal to scale WDQS on FoundationDB (to be deleted soon). I am working on bolder proposal that aims to scale WIKIDATA. Only if you have time, review the page at Grants:Project/Future-proof_WDQS (talk). I will make substantial improvements the next few days, so it is worth watching. I will also move the page to change the title soon. Thanks again for your support and patience. --i⋅am⋅amz3 (talk) 23:24, 26 January 2020 (UTC)Reply

Reset catalogue #3448 edit

My catalogue #3448 worked within the configuration interface, but hasn't produced any results after a few days, and I cannot find a log to indicate what if anything is wrong. I've given it a fairly wide range of numbers to work through, out of which it should be finding around 17,000 entries; it's not clear to me if it is encountering too many errors and failing silently. Would you mind resetting it, please? AndrewNJ (talk) 14:24, 24 March 2020 (UTC)Reply

ListeriaBot on Zu.Wikipedia edit

Hi Magnus, can you please apply for a local bot flag on zuwiki for ListeriaBot or stop running it there? It is flooding recent changes and is editing someones userspace daily for seemingly no reason but it's making it hard to patrol, as userspace patrolling is necessary due to a significant number of spambots and LTAs using this project. Thanks. Praxidicae (talk) 15:16, 23 April 2020 (UTC)Reply

Hi Magnus, it's been several months and I've received no response and the bot is still flooding RC on this very small wiki making it very difficult to patrol. Praxidicae (talk) 15:02, 8 June 2020 (UTC)Reply

Mix 'n' match catalog deactivation request (3521) edit

I made an error in the ID formatting of the catalog and created a new corrected one, can you please deactivate 3521? --Canley (talk) 00:44, 25 April 2020 (UTC)Reply

Done. --Magnus Manske (talk) 14:52, 4 May 2020 (UTC)Reply

Please remove an entry in BaGLAMa2 edit

The category

Files of the State Library of Western Australia

was added but the category name is wrong, so can this entry be deleted please. The correctly named category has been added successfully. Thanks Kerry Raymond (talk) 21:41, 3 May 2020 (UTC)Reply

Also, another with the explicit name

Category:Files from the State Library of Western Australia

Thanks Kerry Raymond (talk) 21:47, 3 May 2020 (UTC)Reply

Both done. --Magnus Manske (talk) 14:50, 4 May 2020 (UTC)Reply

Guide for using GLAMorgan and BaGLAMa2? edit

Hello Magnus,

I hope you are doing well. Could you please help to write a guide for using GLAMorgan and BaGLAMa2? You could write it here or direct me to where I could find it if it has been written already. Thank you for your help. T CellsTalk 21:00, 25 May 2020 (UTC)Reply

BVLarramendi ID and authority control in Wikipedia edit

BVLarramendi ID edit

Dear Magnus I want to thank you for Mix'n'match. In August 2019 the upload of the Polymath Virtual Library catalog to Mix'n'match was completed. Everything went perfectly and almost all Wikidata entries already have the corresponding BVLarramendi ID. However, 10 months later this link does not appear in the authority control space of Wikipedia items (as Template: Authority control say, it would be sufficient that the entries have "authority control" template) Perhaps it is necessary to carry out some other step that I do not know.

Please excuse me if this is not the place for this question, I will be very grateful if you can give me any answer or contact to anybody to consult. Many thanks.

FIST down? edit throws a 404 error. Just in case you did not yet know. Would be cool if it got fixed soon^^ Regards SoWhy 19:57, 19 June 2020 (UTC)Reply

Just go to which redirects you to --Magnus Manske (talk) 17:09, 20 June 2020 (UTC)Reply

Tool:Sighting not working edit

Set of tools "sighting" not working Wargo (talk) 21:38, 12 July 2020 (UTC)Reply

Should be back now. --Magnus Manske (talk) 13:37, 13 July 2020 (UTC)Reply

Fehler CommonsDelinker edit

Moin Magnus! Hier hat dein CommonsDelinkerBot wohl einen Fehler gemacht. Liebe Grüße --Ameisenigel (talk) 20:23, 29 July 2020 (UTC)Reply

Append records to an existing catalogue edit

Dear Magnus, I tried to create a catalogue with a subset of records in CSV, now I would like to add all the records, is it possible? Should I create a new catalogue? Thanks --Luckyz (talk) 12:04, 25 August 2020 (UTC)Reply

Hi, that is not possible via web interface, but if you give me the data, I can do it. Just put it somewhere (pastebin, some free upload service, dropbox etc.) and let me know (best on my Wikidata user page). --Magnus Manske (talk) 12:21, 28 August 2020 (UTC)Reply

Please have a look at T256533 edit

Dear Magnus,

I would like to bring your attention to phab:T256533, as ListeriaBot is currently consuming over 20 % of all login requests. Please be considerate of other users and of the infrastructure. This has been reported to you in [13] already, hence I'm writing this follow-up message.

Please note the block will be blocked if this is not fixed soon.

For more context, please see phab:T256533.

--Martin Urbanec (talk) 18:17, 28 August 2020 (UTC)Reply

Changing range of URLs for Mix n' match scraper edit